Jump to content

Heny

Member
  • Posts

    1
  • Joined

  • Last visited

Reputation Activity

  1. Agree
    Heny got a reaction from Paddo in Thread for Linus Tech Tips Video Suggestions   
    Liquid Metal Thermal Paste as Water Cooling!!

    I know it sounds super stupid but I would love to see the results, would it be like water cooling in steroids?
  2. Like
    Heny reacted to Jose Roberto in Thread for Linus Tech Tips Video Suggestions   
    Hi from Brazil!
     
    I have a suggestion about a simple experimentation on cooling. I've heard several years ago that some people took the extreme measure of sanding the surface of the microprocessor to optimize the heat transfer, making it as smooth as a mirror. Supposedly, this surface treatment would increase the direct contact between the processor and the cooler and then increase the heat transfer.
     
    With this concept in my mind - that a direct contact between both surfaces provides the best possible heat transfer - I began to apply a simpler technique that seems to work. I never really tested not applying this technique to see if it makes any difference, I always apply it, and I think you are the best ones to do experiments on that. If it works, everybody will benefit from it because it's very easy to do.
     
    So, what is this technique?
     
    It's quite simple. The idea in not sanding the surface of the processor nor avoiding the thermal paste. What I do is apply the thermal paste as usual (as little as possible) and, after placing the cooler on the processor, and before tightening the screws, I slide the cooler left and right and back and forth several times (like drawing a cross) while applying some pressure on the cooler against the processor. After that immediately tighten the cooler, preferably while keeping the pressure on the cooler. The effect of the sliding causes is to spread the thermal paste more evenly across the surfaces in contact and expel the excess to the borders. Of course, even with the screws of the cooler loosen, the cooler still does not have too much freedom to move, just some millimeters, but sliding it several times (I didn't count) will reach the desired effect.
     
    To make this technique easier to understand, take an Oreo (the sandwich cookie) and try to squeeze it. It will probably break and the cream in the middle will barely move. Now try squeezing it while sliding sideways. Most of the cream will be expelled from the middle and the wafers will touch each other applying much less pressure. That's the whole idea.
     
    Now, you may want to know what results I've got so far. Well, I'm not a moder nor do experiments on cooling, but what I managed to get was temperatures at about 88°C with the stock cooler of my old Core i7 2600 at 100% CPU in all cores - now I use only a laptop. In contrast, a friend of mine, which had an Intel i7 of the 4 thousands series, said to me he had to buy a special cooler because the stock cooler was not being enough (temperatures going over 100°C). He installed the cooler the way everybody does. Since there are several differences between both setups I cannot ensure that my technique worked so well, so here am I suggesting this to you.
     
    Finally, since we're talking about simple experiments on cooling, if you decide to make a video about this technique, you could test the other techniques too. I would suggest that one about sanding the surface of the processor, and, taking the opportunity that you already have a processor with a surface smooth like that, to test it without the thermal paste.
     
    I hope this suggestion is useful!
     
    Best regards,
     
    José Roberto
  3. Like
    Heny reacted to CPotter in Thread for Linus Tech Tips Video Suggestions   
    Hit us with your best Linus Tech Tips video suggestions! This is to replace our old "What should we review next" thread. Linus or one of the writers will read these suggestions, but they may not reply to you in this thread directly.
     
    Linus Tech Tips https://www.youtube.com/LinusTechTips
×